Abstract
Neonatal cardiac performance is dependent on calcium delivery to the myocardium. Little is known about the use and impact of calcium chloride infusions in neonates who undergo cardiac surgery. We hypothesized that the use of calcium chloride infusions would decrease the doses required of traditional inotropic and vasoactive medications by supporting cardiac output in this patient population. We performed a single-institution, retrospective, cohort study. All neonates (≤ 30 days old) undergoing cardiac surgery from 06/01/2015 through 12/31/2018 were included. Patients were divided into two groups: those who received postoperative calcium chloride infusions (calcium group) and those who did not (control group). The primary outcome was the occurrence of a maximum Vasoactive Inotropic Score (VIS) > 15 in the first 24 h following surgery. One hundred and thirty-five patients met inclusion criteria. Sixty-six patients received postoperative calcium infusions and 69 patients did not. Gestational age, weight at surgery, age at surgery, surgical complexity and cardiopulmonary bypass times were similar between groups. Forty-two (70%) patients receiving calcium had a postoperative maximum VIS > 15 compared with 38 (55%) patients not on a calcium infusion (p = 0.08). There were no differences in postoperative length of ventilation, time to enteral feeding, hospital LOS, or operative mortality between groups. Calcium chloride infusions in neonates who underwent cardiac surgery did not decrease exposure to other inotropic and vasoactive agents in the first 24 post-operative hours or improve patient outcomes.
Keywords: Neonate, Calcium, Congenital heart surgery, Outcomes, Congenital heart disease, Intensive care
Introduction
Calcium homeostasis drives myocardial performance and is regulated by sarcolemmal mechanisms [1]. Compared to the adult myocardium, the neonatal sarcoplasmic reticulum is immature and unable to store and release necessary amounts of calcium to effect contraction, therefore neonatal cardiac performance is highly dependent on calcium delivery to the myocardium [2].
Inotropic medications have long been used to support postoperative myocardial function. Traditional inotropic medications such as dopamine and epinephrine come with unwanted side effects, including tachycardia, increased myocardial oxygen consumption, and increased risk of arrhythmia [3], which may negatively impact the immediate post-operative period for neonates.
Evidence supporting the use of calcium chloride infusions to augment cardiac performance after neonatal cardiac surgery is lacking. A single center, retrospective study found that the use of calcium chloride infusions improved markers of cardiac output in a heterogeneous pediatric population admitted to the cardiac intensive care unit (CICU) [4]. However, this analysis lacked a control or comparison group and neonates who underwent cardiac surgery comprised less than one third of the study population and so information regarding the efficacy of calcium chloride infusions in this specific group is lacking.
The objective of this was study was to determine if calcium chloride infusions improved cardiac performance in neonates who underwent cardiac surgery. We hypothesized that neonates who received calcium chloride infusions would receive lower doses of traditional inotropic and vasoactive medications as measured by the maximum vasoactive inotropic score (VIS) and demonstrate improved markers of cardiac output compared to a control group.
Materials and Methods
Patients
We performed a single-institution, retrospective, cohort study. The study was approved by the Medical University of South Carolina’s (MUSC) Institutional Review Board. All neonates (≤ 30 days old) undergoing cardiac surgery (with and without cardiopulmonary bypass) from 06/01/2015 through 12/31/2018 who received postoperative care in the dedicated pediatric CICU were included. Exclusion criteria included infants who returned from the operating room on Extracorporeal Membrane Oxygenation (ECMO) and infants who underwent isolated patent ductus arteriosus (PDA) ligation in the neonatal intensive care unit.
Internal cardiology and cardiothoracic surgery databases and the electronic health record (EHR) were utilized to collect data that included patient, preoperative, operative, and postoperative variables of interest. Demographic data collected included gestational and chronological age at surgery, birth weight and weight at time of surgery, height, and sex. Chromosomal abnormalities were also noted (specifically trisomy 21 and 22qll.2 deletion syndrome) if a microarray was collected and the result was available in the EMR. Specific cardiac diagnosis was collected, and patients were further classified as having single or biventricular physiology. The operation was documented and categorized according to The Society of Thoracic Surgeons- European Association for Cardiothoracic Surgery (STAT) risk stratification system. Operative data collected included aortic cross-clamp time, cardiopulmonary bypass time (CPB), minimum temperature charted in the operating room, and lowest ionized calcium level in the operating room. During the study period, our institution was conducting a randomized placebo-controlled clinical trial entitled “Corticosteroid Therapy in Neonates Undergoing Cardiopulmonary Bypass,” which has since been published by Graham et al. [5]. Of note, we were able to determine which neonates received intraoperative corticosteroids and control for this potential confounder in the current analysis.
Our institution began routinely using postoperative calcium chloride infusions in neonates undergoing cardiac surgery in January 2017. Prior to adopting empiric initiation of calcium chloride infusions in the operating room, patients were treated as needed with calcium boluses (either calcium gluconate or calcium chloride) at the discretion of the treatment team based on the clinical status of the patient if the ionized calcium level was low (goal 1.2 – 1.4). For our study, patients were divided into 2 groups for comparison. All neonates who underwent cardiac surgery and received a calcium infusion were included in the “calcium group”. All neonates who underwent cardiac surgery and did not receive a calcium infusion were included in the “control group”. Typically, the calcium chloride infusion was started in the operating room at a dose of 5–10 mg/kg/hr. Occasionally it was started upon return to the CICU at the discretion of the medical team. Hemodynamic and laboratory data were collected over the following 24 h and throughout the hospital course. Heart rate (HR), systolic blood pressure (SBP), diastolic blood pressure (DBP), and cSO2 obtained by NIRS monitoring were collected on arrival and at 6, 12, and 24 h postoperatively. Preoperative creatinine was collected on the morning of surgery and postoperative creatinine was monitored for the entirety of the patient stay. Renal injury was defined as a serum creatinine increase during admission greater than 2 times the upper limit of normal (normal range, 0.3–0.7; renal injury > 1.5 mg/dL) [6]. Lactate and serum ionized calcium levels were obtained from arterial blood gases on arrival to the CICU and at 6, 12, and 24 h and lactic acidosis was defined as an increasing arterial lactate concentration > 5 mmol/dL postoperatively [7]. Postoperative aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were collected in most cases and used to define acute liver injury. AST or ALT greater than 2 times the upper limit of normal (normal range, AST 8–40 IU/L; hepatic injury AST > 80 IU/L; normal range, ALT 7–40 IU/L; hepatic injury, ALT > 80 IU/L) at least 24 h after surgery defined hepatic injury [7]. Inotrope dose was collected on arrival to the ICU and at 6, 12, and 24 h, and used to calculate a VIS (vasoactive inotropic score) previously described by Gaies et al. [8, 9]. Our primary outcome was to compare the occurrence of a maximum VIS > 15 in the first 24 h following surgery between the calcium and control groups.
Secondary outcomes included the utilization of ECMO during the postoperative period, incidence of cardiac arrest, duration of initial postoperative mechanical ventilation, evidence of end organ injury, days to initiation of postoperative feeds, hospital length of stay, and operative mortality. A previously described and validated composite outcome was also compared between groups, with patients having met this outcome if they experienced any of the following after cardiac surgery and before hospital discharge: death, circulatory collapse requiring chest compressions (cardiopulmonary resuscitation [CPR]), mechanical circulatory support (ECMO), hepatic insufficiency, renal insufficiency, and lactic acidosis [7].
Statistics
Patient data are reported as mean ± standard deviation for data that are parametric and median (interquartile range) for data that are non-parametric. Differences in means or medians between groups were assessed using independent t tests or Mann–Whitney U tests, respectively. Categorical variables are presented as n (%) and differences between groups were tested using Fisher’s Exact test. Univariable linear and logistic regression analyses were performed to assess associations between outcome and independent variables. Upon univariable regression, independent variables were included in stepwise multivariable regression analyses if their p value at univariable regression was ≤ 0.20. Independent variables were kept in the multivariable model if their p < 0.05, or, p ≤ 0.10 and their inclusion in the model increased the explanatory power of the model by ≥ 0.03 A p value of < 0.05 was considered significant. Statistical analyses were performed using SPSS version 25 (IBM, Armonk, NY).
Results
Patient Characteristics
A total of 135 patients were included in the analysis with 66 patients in the calcium group and 69 patients in the control group. Patient characteristics are shown in Table 1. There were fewer females in the calcium group (52% vs 33%, p = 0.03). The groups were otherwise similar with regard to age at surgery (7 vs 8 days, p = 0.18), gestational age at time of surgery (39 vs 38 weeks, p = 0.09), weight at surgery (3.3 vs 3.2 kg, p = 0.40), and presence of a chromosomal abnormality (23% vs 14%, p = 0.22). There were no differences observed between groups with respect to single ventricle operations (44% vs 44%, p = 0.96), arch repair operations (44% vs 42%, p = 0.82), or surgical complexity as defined by number of patients in STAT categories 4 or 5 (79% vs 77%, p = 0.78). More patients in the calcium group underwent aortic cross-clamp during surgery (83% vs 58%, p < 0 0.01). Additionally, more patients in the calcium group received intraoperative steroids (47% vs 15%, p = < 0.01).
Table 1.
Patient characteristics
| Calcium Group (n = 66) | Control Group (n = 69) | p value | |
|---|---|---|---|
| Age at surgery (days) | 7 (6, 10) | 8 (6, 11) | 0.18 |
| Gestational age at surgery (weeks) | 39 (38, 39) | 38 (37, 39) | 0.09 |
| Female sex | 22 (33%) | 36 (52%) | 0.03 |
| Weight at surgery (kilograms) | 3.3 ± 0.5 | 3.2 ± 0.6 | 0.40 |
| Chromosomal abnormality | 15 (23%) | 10 (14%) | 0.22 |
| 22q11 deletion syndromea | 2 (4%) | 2 (4%) | 1.0 |
| Surgical procedure group | |||
| Single ventricle | 29 (44%) | 30 (44%) | 0.96 |
| Arch repair | 29 (44%) | 29 (42%) | 0.82 |
| STAT category | |||
| 4 or 5 | 52 (79%) | 53 (77%) | 0.78 |
| CPB utilized | 60 (91%) | 48 (70%) | < 0.01 |
| CPB time (minutes) | 187 ± 66 | 177 ± 64 | 0.42 |
| Aortic cross-clamp occurred | 55 (83%) | 40 (58%) | < 0.01 |
| Lowest ionized calcium in OR (prior to starting infusion) | 0.71 ± 0.22 | 0.83 ± 0.28 | 0.01 |
| Calcium infusion initiated in OR | 58 (94%) | NA | NA |
| Received intraoperative steroids | 29 (47%) | 10 (15%) | < 0.01 |
| Received postoperative steroids | 24 (39%) | 23 (33%) | 0.48 |
| Delayed sternal closure | 31 (51%) | 26 (38%) | 0.15 |
| Calcium infusion duration (hours) | 37 (20, 63) | NA | NA |
Categorical data are presented as n (%). Continuous data are presented as mean + / − SD for data that are normally distributed or median (interquartile range) for data that are not normally distributed
STAT The Society of Thoracic Surgeons- European Association for Cardiothoracic Surgery, CPB cardiopulmonary bypass, OR operating room, NA not applicable
Denominator equals number of patients who underwent genetic testing (n = 50 for each group)
Primary Outcome
A greater proportion of patients in the calcium group developed a maximum VIS > 15 in the first 24 h following surgery but this finding did not reach statistical significance (70% vs 55%, p = 0.08). The VIS for both groups at each time interval is shown in Fig. 1. The calcium group had a higher maximum VIS in the first 24 h following surgery (17.3 vs 14.9, p = 0.04). However, upon multivariable regression analysis, calcium infusion was not independently associated with maximum VIS > 15 (p = 0.64). The only independent predictor of maximum VIS > 15 was whether the patient underwent aortic cross-clamp during surgery (OR 4.9 (95% CI 2.2–11.0), p < 0.01).
Fig. 1.
Vasoactive inotropic scores for the calcium (red) and control groups (blue)
Secondary Outcomes
Markers of cardiac output including systolic and diastolic blood pressure, heart rate, lactate, and cerebral NIRS are presented in Table 2. There were no differences observed in these parameters between the study groups except for higher lactate levels in the calcium group upon admission to the CICU and at 6- and 12-h following surgery (p = 0.01). Lactate levels were similar between groups at 24 h following surgery (p = 0.65). Upon multivariable regression, calcium infusion was not predictive of lactate levels. There was no difference in the rate of postoperative cardiac arrest, duration of mechanical ventilation, time to feeding initiation, or hospital length of stay, or the composite outcome between the groups as per Table 2.
Table 2.
Impact of calcium chloride infusions on postoperative outcomes, univariate analysis
| Calcium Group (n = 66) | Control Group (n = 69) | p value | |
|---|---|---|---|
| Systolic blood pressure | |||
| Admission to ICU | 81 (72, 94) | 80 (72, 90) | 0.45 |
| 6 h | 70 (63, 77) | 72 (66, 79) | 0.25 |
| 12 h | 70 (65, 75) | 73 (66, 78) | 0.12 |
| 24 h | 71 (68, 77) | 73 (67, 82) | 0.40 |
| Diastolic blood pressure | |||
| Admission to ICU | 49 (43, 57) | 47 (40, 55) | 0.41 |
| 6 h | 46 (42, 50) | 47 (43, 51) | 0.72 |
| 12 h | 45 (40, 51) | 45 (40, 51) | 0.97 |
| 24 h | 43 (39, 49) | 45 (39, 50) | 0.84 |
| Heart rate | |||
| Admission to the ICU | 157 (150, 167) | 161 (148, 176) | 0.27 |
| 6 h | 169 (150, 176) | 165 (150, 179) | 0.95 |
| 12 h | 167 (157, 176) | 169 (151, 179) | 0.62 |
| 24 h | 157 (149, 171) | 159 (147, 168) | 0.54 |
| Cerebral NIRS | |||
| Admission to the ICU | 53 ± 12 | 55 + 16 | 0.48 |
| 6 h | 55 ± 12 | 57 ± 14 | 0.61 |
| 12 h | 59 ± 12 | 58 ± 13 | 0.61 |
| 24 h | 64 ± 13 | 63 ± 14 | 0.85 |
| Lactate | |||
| Admission to the ICU | 2.6 (2.0, 3.6) | 2.0 (1.5, 3.3) | 0.01 |
| 6 h | 2.5 (1.4, 3.3) | 1.5 (1.1, 2.4) | < 0.01 |
| 12 h | 2.3 (1.6, 3.0) | 1.7 (1.2, 2.4) | < 0.01 |
| 24 h | 1.5 (1.0, 2.0) | 1.6 (1.0, 2.3) | 0.65 |
| Vasoactive Inotropic Score (VIS) | |||
| Admission to the ICU | 10.5 + 4.7 | 10.1 ± 6.1 | 0.70 |
| 6 h | 13.6 ± 5.9 | 12.4 ± 6.8 | 0.28 |
| 12 h | 12.6 ± 6.1 | 12.9 ± 6.8 | 0.28 |
| 24 h | 13.1 ± 5.8 | 11.5 ± 6.7 | 0.16 |
| Max | 17.3 ± 6.0 | 14.9 ± 7.3 | 0.04 |
| VIS > 10 | 56 (90%) | 51 (74%) | 0.02 |
| CPB patients onlya | 53 (88%) | 44 (92%) | 0.75 |
| Maximum VIS > 15 | 42 (70%) | 38 (55%) | 0.08 |
| CPB patients onlyb | 39 (65%) | 33 (69%) | 0.84 |
| Renal injuryc | 2 (3%) | 3 (4%) | 1.0 |
| Acute liver injuryd | 6 (10%) | 5 (8%) | 0.76 |
| ECMO | 5 (8%) | 8 (12%) | 0.57 |
| Cardiac arrest | 5 (8%) | 6 (9%) | 1.0 |
| Duration of postoperative ventilation (days) | 4 (3, 7) | 4 (2, 8) | 0.93 |
| Time to postoperative feed initiation (days) | 4 (3, 5) | 3 (2, 5) | 0.11 |
| Hospital LOS (days) | 37 (25, 53) | 33 (21, 54) | 0.29 |
| Operative mortality | 4 (6%) | 8 (12%) | 0.37 |
| Composite outcome | 24 (39%) | 24 (35%) | 0.72 |
Categorical data are presented as n (%). Continuous data are presented as mean (standard deviation) for data that are normally distributed or median (interquartile range) for data that are not normally distributed
CPB cardiopulmonary bypass, ICU intensive care unit, NIRS near-infrared spectroscopy, VIS vasoactive inotropic score, ECMO extracorporeal membrane oxygenation, LOS length of stay
Denominator for patients who received CPB and calcium infusion = 60
Denominator for patients who received CPB and no calcium infusion = 48
Denominator for renal injury and calcium group = 62, denominator for control group = 69
Denominator for acute liver injury and calcium group = 60, denominator for control group = 63
Conclusions
To our knowledge, this is the largest analysis describing the association of calcium chloride infusions with clinical outcomes in neonates who underwent cardiac surgery. In this cohort, the use of calcium chloride infusions was not associated with decreased use of other inotropic medications in the early postoperative period and did not significantly affect markers of cardiac output.
The findings of this study contrast with those reported by Averin et al. [4]. In that study, patients who received calcium chloride infusions demonstrated improved markers of cardiac output. The population ranged from 0–17 years in age and included both surgical and non-surgical patients, thus calcium chloride infusions were not always initiated for the sole purpose of postoperative support. Neonates demonstrated the most significant improvement in hemodynamics. The study populations differed significantly between that study and our analysis which may account for the discrepant findings. Additionally, the lack of a comparison or control group in the study by Averin et al. limits the ability to attribute hemodynamic improvement to the calcium chloride infusion [4].
While our calcium and control study groups were similar with respect to surgical acuity as measured by STAT category, one potentially important difference was the more frequent occurrence of aortic cross clamping during surgery in the calcium group (83%) compared with patients in the control group (58%). However, there remained no association between the use of calcium chloride infusions and maximum VIS when aortic cross clamping was controlled for in multivariable analysis. Lastly, there was no association identified between calcium infusions and maximum VIS in a subgroup analysis that excluded patients who did not receive cardiopulmonary bypass (65% vs 69%, p = 0.84). It is possible that unmeasured differences existed between groups with respect to operative techniques, anesthetic management or blood product resuscitation and could have impacted the primary outcome of this study. With that being acknowledged, the same two surgeons performed all of the operations and there were no major changes with regard to intraoperative care during the study period that was obvious to the investigators other than increasing use of steroids prior to initiation of cardiopulmonary bypass.
Importantly, the impact of calcium chloride infusions on the post-bypass period within the operating room was not studied. Anecdotally, our anesthesia and perfusion colleagues have noticed less frequent hypocalcemia (often associated with blood product resuscitation) since this practice has been adopted. Attention to whether calcium chloride infusions have influenced care in the operating room after separation from bypass is an opportunity for further study.
This study has several limitations. The retrospective nature of the study limits our ability to understand if post-operative decision-making in the CICU and blood pressure goals were similar between groups. The blood pressures analyzed between groups were abstracted from the EHR at specific time points and thus may not reflect fluctuations in continuous arterial blood pressure monitoring (and concurrent vasoactive or inotropic drug dose adjustments) inherent to the early postoperative period of neonatal cardiac surgery. Although most calcium chloride infusions were initiated in the operating room when the patient separated from cardiopulmonary bypass, it is possible that some infusions were started due to high or increasing inotrope requirements. This practice may confound the study results. Lastly, the use of a “historical” control group may have resulted in an era effect despite no change in cardiac surgeons during the study period.
In conclusion, we found the use of calcium chloride infusions was not associated with lower inotropic or vasoactive requirements or clinical outcomes in neonates who underwent cardiac surgery. Given the findings of this study, it is reasonable to consider a more selective, targeted approach to the use of calcium chloride infusions. While there was no obvious harm identified in patients who received calcium chloride infusions, there are compatibility issues with other medications that can result in the formation of a precipitate within catheter lumens [10]. Future directions should include investigation of the impact of calcium chloride infusions on care delivered within the operating room after separation from cardiopulmonary bypass.
Funding
No funding was provided for this study, and no honorarium or other form of payment was provided to anyone to produce the manuscript.
Footnotes
Code Availability Not applicable.
Conflict of interest The authors have no conflicts of interest to disclose.
Data Availability
The authors agree to data transparency.
References
- 1.Morgan JP (1991) Abnormal intracellular modulation of calcium as a major cause of cardiac contractile dysfunction. N Engl J Med 325(9):625–632. 10.1056/NEJM199108293250906 [DOI] [PubMed] [Google Scholar]
- 2.Balaguru D, Haddock PS, Puglisi JL, Bers DM, Coetzee WA, Artman M (1997) Role of the sarcoplasmic reticulum in contraction and relaxation of immature rabbit ventricular myocytes. J Mol Cell Cardiol 10:2747–2757. 10.1006/jmcc.1997.0509 [DOI] [PubMed] [Google Scholar]
- 3.Hoffman TM, Wernovsky G, Atz AM, Kulik TJ, Nelson DP, Chang AC, Bailey JM, Akbary A, Kocsis JF, Kaczmarek R, Spray TL, Wessel DL (2003) Efficacy and safety of milrinone in preventing low cardiac output syndrome in infants and children after corrective surgery for congenital heart disease. Circulation 107(7):996–1002. 10.1161/01.cir.0000051365.81920.28 [DOI] [PubMed] [Google Scholar]
- 4.Averin K, Villa C, Krawczeski CD, Pratt J, King E, Jefferies JL, Nelson DP, Cooper DS, Ryan TD, Sawyer J, Towbin JA, Lorts A (2016) Initial observations of the effects of calcium chloride infusions in pediatric patients with low cardiac output. Pediatr Cardiol 37(3):610–617. 10.1007/s00246-015-1322-2 [DOI] [PubMed] [Google Scholar]
- 5.Graham EM, Martin RH, Buckley JR, Zyblewski SC, Kavarana MN, Bradley SM, Alsoufi B, Mahle WT, Hassid M, Atz AM (2019) Corticosteroid therapy in neonates undergoing cardiopulmonary bypass: randomized controlled trial. J Am Coll Cardiol 74(5):659–668. 10.1016/j.jacc.2019.05.060 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Butts RJ, Scheurer MA, Zyblewski SC, Wahlquist AE, Nietert PJ, Bradley SM, Atz AM, Graham EM (2014) A composite outcome for neonatal cardiac surgery research. J Thorac Cardiovasc Surg 147(1):428–433. 10.1016/j.jtcvs.2013.03.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Charpie JR, Dekeon MK, Goldberg CS, Mosca RS, Bove EL, Kulik TJ (2000) Serial blood lactate measurements predict early outcome after neonatal repair or palliation for complex congenital heart disease. J Thorac Cardiovasc Surg 120(1):73–80. 10.1067/mtc.2000.106838 [DOI] [PubMed] [Google Scholar]
- 8.Gaies MG, Gurney JG, Yen AH, Napoli ML, Gajarski RJ, Ohye RG, Charpie JR, Hirsch JC (2010) Vasoactive-inotropic score as a predictor of morbidity and mortality in infants after cardiopulmonary bypass. Pediatr Crit Care Med 11(2):234–238. 10.1097/PCC.0b013e3181b806fc [DOI] [PubMed] [Google Scholar]
- 9.Gaies MG, Jeffries HE, Niebler RA, Pasquali SK, Donohue JE, Yu S, Gall C, Rice TB, Thiagarajan RR (2014) Vasoactive-inotropic score is associated with outcome after infant cardiac surgery: an analysis from the Pediatric Cardiac Critical Care Consortium and Virtual PICU System Registries. Pediatr Crit Care Med 15(6):529–537. 10.1097/PCC.0000000000000153 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Weeks PA, Teng Y, Wu L, Sun M, Yang Z, Chow DS (2014) Chemical and physical compatibility of an intravenous solution of epinephrine with calcium chloride. Int J Pharm Compd 18(2):152–158 [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The authors agree to data transparency.